Cellular structures coarsen according to a local evolution law, a gradient flow or curvature driven growth, for example, limited by space filling constraints, which give rise to random changes in configuration. Composed of volumes, facets, their boundaries, and so forth, they are ensembles of singlular structures. Among the most challenging and ancient of such systems are polycrystalline granular networks, especially those which are anisotropic, ubiquitous among engineered materials. It is the problem of microstructure. These are large scale metastable, active across many scales. We discuss recent work in this area, especially the discovery and the theory of the GBCD, the grain boundary character distribution, which offers promise as a predictive measure of texture related material properties. There are many mathematical challenges and the hint of universality.